Data Viz Project - 200
Some Data Cleaning.
Just in case 100 file is old or more general help check
#Look for hourly patterns
hourly_data <- coc
# Clean up bad column names
hourly_data <- hourly_data %>% clean_names()
#na.omit(hourly_data)
#summary(hourly_data)
hourly_data$hour <- hour(as.POSIXlt(hourly_data$trip_start_timestamp, format="%m/%d/%Y %I:%M:%S %p")) Look at the tip mean. 64.4 cents!!
Look into Tip Information
Yes, I’m going to make a pie chart below. It will highlight that only 20% of riders tip. 80% leave no tip! This is the only time I would use a pie chart. Showing two or three numbers that are out of balance or equal weights.
This was harder than I would have guested. I had to do a lot of numberic conversions below. Adding the title must have taken 1 hour or more.
tips <- subset(hourly_data, hourly_data$tip > 0)
hourly_data$isTipper <- FALSE
hourly_data$isTipper[hourly_data$tip > 0] <- TRUE
my_pie <- data.frame(
group=c("Tippers", "NoTips"),
value2=c(toString(count(tips)),toString(count(hourly_data)-count(tips)))
)
my_pie$value <- as.numeric(as.character(my_pie$value))
my_pie <- my_pie[,-c(2)]
head(my_pie)## group value
## 1 Tippers 9961
## 2 NoTips 42039#ggplot(my_pie, aes(x="", y=value, fill=group)) +
# geom_bar(stat="identity", width=1, color="white") +
# coord_polar("y", start=0) +
# theme_void()
df <- my_pie %>%
# factor levels need to be the opposite order of the cumulative sum of the values
mutate(group = group,
cumulative = cumsum(value),
midpoint = cumulative - value / 2,
label = paste0(group, " ", round(value / sum(value) * 100, 1), "%"))
ggplot(df, aes(x = 1, weight = value, fill = group)) +
geom_bar(width = 1, position = "stack") +
coord_polar(theta = "y") +
geom_text(aes(x = 1.3, y = midpoint, label = label)) +
theme(axis.ticks = element_blank(),
axis.text = element_blank(),
axis.title = element_blank(),
plot.caption = element_text(hjust = 0.5)) +
ggtitle("Percentage of Riders Who Tip") 
#general pie chart - no ggplot
#pie(my_pie$value, my_pie$group)
percTip <- count(tips) / (count(hourly_data))
percTip## n
## 1 0.1915577Above we show most pepople don’t tip.
Let’s look at those that do tip. How are they doing percentage wise?
fare vs tip - what % are drivers getting?
ggplot (hourly_data, aes(x=hourly_data$fare, y=tip)) +geom_point()
#### look at historgram without zero tips.
ggplot (tips, aes(x=tips$tip)) +geom_histogram(bins = 24)
#### First look at historgram without zero tips.
ggplot (hourly_data, aes(x=hourly_data$shared_trip_authorized, y=tip))+ geom_boxplot()
Look into rides by hour
Histogram of Rides by Hour below
ggplot (hourly_data, aes(x=hourly_data$hour)) +geom_histogram(bins = 24)
#ggplot (hourly_data, aes(x=hourly_data$trip_start_timestamp, y=`trip_total`))+ geom_col()Pick-Up Density
Now let’s look at all the pick-up locations:
# Chicago = 41.8781° N, -87.6298° W
p <-ggmap(get_googlemap(center = c(lon = -87.629800, lat =41.878100 ),
zoom = 11, scale = 2,
maptype ='roadmap',
color = 'color', key = myKey))## Source : https://maps.googleapis.com/maps/api/staticmap?center=41.8781,-87.6298&zoom=11&size=640x640&scale=2&maptype=roadmap&key=xxxp + geom_point(aes(x = Pickup.Centroid.Longitude, y = Pickup.Centroid.Latitude), data = coc, size = 0.7) +
theme(legend.position="bottom")## Warning: Removed 5918 rows containing missing values (geom_point).
Drop-Off Density
Now let’s look at all the drop-off locations:
d <-ggmap(get_googlemap(center = c(lon = -87.629800, lat =41.878100 ),
zoom = 11, scale = 2,
maptype ='terrain',
color = 'color', key = myKey))## Source : https://maps.googleapis.com/maps/api/staticmap?center=41.8781,-87.6298&zoom=11&size=640x640&scale=2&maptype=terrain&key=xxxd + geom_point(aes(x = Dropoff.Centroid.Longitude, y = Dropoff.Centroid.Latitude), data = coc, size = 0.7) +
theme(legend.position="bottom")## Warning: Removed 6710 rows containing missing values (geom_point).
######### #REPEAT ABOVE WITH ZOOM ######### ### Pick-Up Density
Now let’s look at all the pick-up locations:
# Chicago = 41.8781° N, -87.6298° W
p <-ggmap(get_googlemap(center = c(lon = -87.629800, lat =41.878100 ),
zoom = 12, scale = 2,
maptype ='terrain',
color = 'color', key = myKey))## Source : https://maps.googleapis.com/maps/api/staticmap?center=41.8781,-87.6298&zoom=12&size=640x640&scale=2&maptype=terrain&key=xxxp + geom_point(aes(x = Pickup.Centroid.Longitude, y = Pickup.Centroid.Latitude), data = coc, size = 0.7) +
theme(legend.position="bottom")## Warning: Removed 15497 rows containing missing values (geom_point).
Drop-Off Density
Now let’s look at all the drop-off locations:
d <-ggmap(get_googlemap(center = c(lon = -87.629800, lat =41.878100 ),
zoom = 14, scale = 2,
maptype ='terrain',
color = 'color', key = myKey))## Source : https://maps.googleapis.com/maps/api/staticmap?center=41.8781,-87.6298&zoom=14&size=640x640&scale=2&maptype=terrain&key=xxxd + geom_point(aes(x = Dropoff.Centroid.Longitude, y = Dropoff.Centroid.Latitude), data = coc, size = 0.7) +
theme(legend.position="bottom")## Warning: Removed 38094 rows containing missing values (geom_point).
It is a bit tricky to see the density of the pickup and dropoffs because all the points are sitting on top of each other. Below sets the alpha variable which will make the dots transparent. This helps display the density of points plotted.
p + geom_point(aes(x = Pickup.Centroid.Longitude, y = Pickup.Centroid.Latitude), colour = '#011f4b', data = coc, alpha=0.25, size = 0.5) +
theme(legend.position="none")## Warning: Removed 15497 rows containing missing values (geom_point).
p + stat_density2d(
aes(x = Pickup.Centroid.Longitude, y = Pickup.Centroid.Latitude,
fill = ..level.., alpha = 0.15), size = 0.01, bins = 30,
data = coc, geom = "polygon") #+## Warning: Removed 15497 rows containing non-finite values (stat_density2d).
#geom_point(aes(x = x, y = y, stroke = 2), colour=col4, data = n, size =1.5) +
#geom_label_repel(aes(x, y, label = label), data=n, family = 'Times', size = 3, box.padding = 0.2, point.padding = 0.3, segment.color = 'grey50') p + stat_density2d(
aes(x = Pickup.Centroid.Longitude, y = Pickup.Centroid.Latitude, fill = ..level.., alpha = 0.25),
size = 0.1, bins = 40, data = coc,
geom = "polygon"
) +
geom_density2d(data = coc,
aes(x = Pickup.Centroid.Longitude, y = Pickup.Centroid.Latitude), size = 0.3)## Warning: Removed 15497 rows containing non-finite values (stat_density2d).
## Warning: Removed 15497 rows containing non-finite values (stat_density2d).
Move this to user interaction shiny application
I’m going to move this code to a shiny app so one can see the map change by selecting an hour of the day.
#END